Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for displaying a three-dimensional model image of a portion of a target object, comprising: (a) carrying an imaging device from an initial location that is known in a frame of reference of the target object to a subsequent location that is not known in the frame of reference of the target object while walking along a path, wherein an area on the target object including a part of interest is in a field-of-view of the imaging device when the imaging device is at the subsequent location; (b) acquiring orientation data representing an orientation of the imaging device during walking along the path using an inertial measurement unit incorporated in the imaging device; (c) acquiring vertical acceleration data during walking along the path using the inertial measurement unit; (d) counting a number of walking steps taken during walking along the path by detecting a repeating characteristic of the vertical acceleration data; (e) calculating a current position and a current orientation for the subsequent location of the imaging device relative to the frame of reference of the target object using the orientation data, the number of walking steps, and a step length; and (f) displaying a three-dimensional model image of at least a portion of the target object with a viewpoint that is a function of the calculated current position and current orientation of the imaging device.
This invention relates to three-dimensional (3D) model generation and display, specifically addressing the challenge of accurately reconstructing and visualizing 3D models of objects when the precise position and orientation of the imaging device are not initially known. The method involves an imaging device, equipped with an inertial measurement unit (IMU), being moved from a known starting position to an unknown subsequent position while capturing images of a target object. During this movement, the IMU records the device's orientation and vertical acceleration. The vertical acceleration data is analyzed to detect a repeating pattern indicative of walking steps, allowing for a count of the steps taken. Using the acquired orientation data, the counted number of steps, and an estimated step length, the method calculates the imaging device's precise position and orientation relative to the target object's frame of reference at the subsequent location. Finally, a 3D model of at least a portion of the target object is displayed. The viewpoint of this displayed model is determined by the calculated position and orientation of the imaging device, effectively reconstructing the scene from the device's perspective as it moved.
2. The method as recited in claim 1 , further comprising: selecting a point in a portion of the displayed three-dimensional model image that depicts the part of interest; retrieving part identification data from a three-dimensional model database corresponding to the part of interest selected; and displaying alphanumeric symbology that depicts the identity of a part contained in the retrieved part identification data.
This invention relates to a system for identifying parts in a three-dimensional (3D) model, particularly in applications such as manufacturing, maintenance, or assembly where visual identification of components is necessary. The problem addressed is the difficulty in quickly and accurately identifying specific parts within complex 3D models, which can be time-consuming and error-prone when done manually. The method involves displaying a 3D model image of an assembly or structure containing multiple parts. A user selects a point within a portion of the displayed 3D model that corresponds to a part of interest. The system then retrieves part identification data from a 3D model database based on the selected point. This data includes alphanumeric symbology, such as part numbers, names, or other identifiers, which is then displayed to the user. The system ensures that the correct part is identified by correlating the selected point with the corresponding part in the database. The method may also include additional steps such as highlighting the selected part in the 3D model to confirm the correct identification or providing additional details about the part, such as specifications or maintenance instructions. The system improves efficiency by automating part identification, reducing the need for manual searches or external documentation. This is particularly useful in industries where quick and accurate part identification is critical, such as aerospace, automotive, or industrial manufacturing.
3. The method as recited in claim 1 , further comprising: acquiring feature image data while the area on the target object including the part of interest is in the field-of-view of the imaging device; displaying a feature image corresponding to the acquired feature image data as an overlay or background to the three-dimensional model image; and adjusting the viewpoint until the three-dimensional model image being displayed matches the feature image.
This invention relates to a system for aligning a three-dimensional (3D) model of a target object with real-time imaging data to enhance visualization and analysis of a specific part of interest. The problem addressed is the difficulty in accurately correlating a pre-existing 3D model with live imaging data, particularly when the model must be aligned with a specific region of the target object. The method involves capturing feature image data of an area on the target object that includes the part of interest while the area is within the field-of-view of an imaging device. The captured feature image is then displayed as an overlay or background to the 3D model image. The viewpoint of the 3D model is adjusted until the displayed 3D model image matches the feature image, ensuring precise alignment between the model and the real-world object. This alignment process allows for accurate visualization and analysis of the part of interest in the context of the 3D model. The method may also include generating the 3D model image by rendering the 3D model from a viewpoint corresponding to the imaging device's position and orientation. Additionally, the feature image data may be acquired using a camera or other imaging device, and the alignment process may involve manual or automated adjustments to the viewpoint. The system ensures that the 3D model accurately represents the target object in real-time, improving applications such as medical imaging, industrial inspection, or augmented reality.
4. The method as recited in claim 3 , wherein the displayed feature image comprises an outline extracted from the feature image data using edge detection.
This invention relates to image processing techniques for enhancing visual features in digital images. The problem addressed is the difficulty in clearly identifying and displaying specific features within an image, particularly when the features are subtle or obscured by surrounding elements. The solution involves extracting and displaying an outline of a feature from the image data using edge detection algorithms. Edge detection is a computer vision technique that identifies boundaries within an image by analyzing pixel intensity gradients. By applying this method, the system isolates the outline of the feature, making it more distinct and easier to perceive. This approach is particularly useful in applications such as medical imaging, where subtle anatomical structures need to be highlighted, or in industrial inspections, where defects or features must be clearly identified. The extracted outline can be overlaid on the original image or displayed separately to emphasize the feature of interest. The method improves feature visibility without altering the underlying image data, preserving the original context while enhancing the desired feature. This technique can be integrated into various imaging systems, including cameras, scanners, and software applications, to provide clearer and more accurate visual representations of features within images.
5. The method as recited in claim 1 , wherein step (e) comprises computing an offset transformation matrix that represents position and orientation offsets of the subsequent location relative to the initial location.
This invention relates to a method for determining positional and orientational relationships between locations in a spatial environment, particularly for applications in robotics, augmented reality, or navigation systems. The method addresses the challenge of accurately tracking changes in position and orientation between an initial location and a subsequent location, which is critical for tasks such as object localization, path planning, or environment mapping. The method involves capturing sensor data at both the initial and subsequent locations, where the sensor data may include images, depth measurements, or other spatial data. The method then processes this data to compute a transformation matrix that quantifies the positional and orientational differences between the two locations. This transformation matrix is derived by aligning the sensor data from the subsequent location with the initial location, accounting for any shifts or rotations that have occurred. The computed matrix provides a precise mathematical representation of the relative movement, enabling systems to update their spatial awareness or adjust their operations accordingly. The transformation matrix is particularly useful in dynamic environments where real-time adjustments are necessary, such as in autonomous navigation or augmented reality applications. By accurately modeling the positional and orientational offsets, the method ensures that subsequent actions or calculations are based on correct spatial relationships, improving system performance and reliability. The method may be implemented using various computational techniques, including optimization algorithms or machine learning models, to enhance accuracy and efficiency.
6. The method as recited in claim 5 , wherein the position offset is determined using a dead-reckoning process.
A system and method for determining the position offset of a mobile device or vehicle involves using a dead-reckoning process to estimate positional changes over time. Dead-reckoning is a navigation technique that calculates current position based on a previously known position, estimated speeds, and course over elapsed time, typically using inertial sensors or wheel rotation data. This method compensates for inaccuracies in global positioning systems (GPS) or other external positioning signals, particularly in environments where such signals are unreliable or unavailable, such as urban canyons, tunnels, or indoor spaces. The dead-reckoning process integrates sensor data, such as accelerometers, gyroscopes, or odometers, to track movement and compute positional offsets relative to a reference point. The system may also incorporate error correction mechanisms, such as periodic recalibration with available GPS signals or other reference points, to improve accuracy. This approach ensures continuous position tracking even when external positioning aids are limited, enhancing navigation reliability in challenging environments. The method is particularly useful in autonomous vehicles, robotics, and mobile devices where precise location data is critical for navigation and operation.
7. The method as recited in claim 5 , wherein the position offset is determined by executing a dead-reckoning algorithm that combines the orientation data from the inertial measurement unit with walking step information to produce a piecewise linear approximation of the path for relative motion measurement.
This invention relates to a method for determining a position offset in a navigation system, particularly for tracking relative motion using inertial measurement data. The method addresses the challenge of accurately estimating position changes when precise global positioning is unavailable, such as indoors or in environments with signal interference. The method involves using an inertial measurement unit (IMU) to capture orientation data, which is then combined with walking step information to compute a position offset. A dead-reckoning algorithm processes this data to generate a piecewise linear approximation of the path traveled. The walking step information may include step detection, step length estimation, and step direction, which are derived from IMU sensor outputs like accelerometers and gyroscopes. The dead-reckoning algorithm integrates these inputs to model the user's movement as a series of linear segments, accounting for changes in orientation and step parameters over time. This approach improves upon traditional dead-reckoning techniques by incorporating detailed step data, enhancing accuracy in tracking relative motion. The method is particularly useful in applications requiring continuous position updates without relying on external positioning signals, such as indoor navigation or wearable tracking devices. The piecewise linear approximation simplifies the computation while maintaining sufficient precision for many practical use cases.
8. The method as recited in claim 1 , wherein step (d) comprises detecting successive vertical acceleration maxima above a specified threshold in the vertical acceleration data, the count being incremented by unity each time a vertical acceleration maximum is detected.
A method for analyzing vertical acceleration data to detect and count successive vertical acceleration maxima above a specified threshold. The method addresses the need to accurately identify and quantify repetitive vertical movements or impacts, such as those occurring in industrial machinery, transportation systems, or biomechanical applications, where monitoring such events is critical for performance, safety, or maintenance purposes. The method involves processing vertical acceleration data to detect peaks or maxima in the signal. Each time a vertical acceleration maximum exceeds a predefined threshold, the system increments a counter by one. This allows for tracking the frequency of significant vertical acceleration events, which can indicate mechanical stress, impact events, or other dynamic conditions requiring attention. The method ensures reliable detection by focusing on successive maxima, meaning it only counts peaks that occur after a previous peak has been identified and recorded. This prevents redundant counting of closely spaced or overlapping peaks, improving accuracy in scenarios where rapid or continuous vertical movements occur. The threshold setting allows for customization based on the specific application, ensuring sensitivity to relevant events while filtering out noise or minor fluctuations. By providing a count of vertical acceleration maxima above the threshold, the method enables real-time or post-processing analysis of dynamic systems, facilitating predictive maintenance, performance optimization, or safety monitoring. The approach is particularly useful in applications where repetitive vertical forces or impacts are a concern, such as in vehicle suspension systems, industrial equipment, or human motion analysis.
9. The method as recited in claim 8 , wherein step (e) comprises using orientation data acquired at instants in time when the successive vertical acceleration maxima above a specified threshold occurred.
This invention relates to a method for analyzing vertical acceleration data to determine the orientation of a moving object, such as a vehicle or a mobile device. The problem addressed is accurately detecting and processing vertical acceleration peaks to improve orientation tracking, particularly in dynamic environments where motion artifacts can distort measurements. The method involves acquiring vertical acceleration data from an accelerometer over time. The data is processed to identify successive vertical acceleration maxima that exceed a specified threshold. These maxima represent significant motion events, such as impacts or changes in direction. Orientation data, such as angular position or tilt, is then acquired specifically at the instants when these vertical acceleration maxima occur. By synchronizing orientation measurements with these key acceleration events, the method reduces noise and improves the accuracy of orientation tracking. The method may also include filtering the vertical acceleration data to remove noise or irrelevant fluctuations before identifying the maxima. The specified threshold ensures that only meaningful acceleration events are considered, enhancing the reliability of the orientation data. This approach is particularly useful in applications where precise orientation tracking is required, such as in vehicle dynamics monitoring, sports performance analysis, or wearable device motion tracking. The synchronization of orientation data with vertical acceleration peaks provides a more robust and accurate representation of the object's motion state.
10. The method as recited in claim 1 , wherein step (d) comprises detecting successive vertical acceleration minima below a specified threshold in the vertical acceleration data, the count being incremented by unity each time a vertical acceleration minimum is detected.
This invention relates to a method for analyzing vertical acceleration data to detect specific motion patterns, particularly in the context of monitoring physical activity or movement. The method addresses the challenge of accurately identifying repetitive vertical movements, such as steps or impacts, by analyzing vertical acceleration signals to distinguish meaningful events from noise. The method involves processing vertical acceleration data to detect successive vertical acceleration minima that fall below a specified threshold. Each time a minimum is detected, a counter is incremented by one. This approach helps quantify the number of significant vertical movements, such as steps taken during walking or running, by filtering out minor fluctuations that do not represent meaningful motion. The threshold ensures that only relevant acceleration minima are counted, improving the accuracy of motion detection. The method may be part of a broader system for activity monitoring, where vertical acceleration data is collected from a wearable or embedded sensor. By focusing on vertical acceleration minima, the method effectively isolates key motion events, making it useful for applications like fitness tracking, gait analysis, or impact monitoring. The threshold-based detection ensures robustness against sensor noise and minor vibrations, enhancing reliability in real-world conditions. This technique is particularly valuable in scenarios where precise motion counting is required, such as in sports performance analysis or medical rehabilitation.
11. The method as recited in claim 10 , wherein step (e) comprises using orientation data acquired at instants in time when the successive vertical acceleration minima below a specified threshold occurred.
This invention relates to a method for analyzing motion data, particularly vertical acceleration data, to determine the orientation of a moving object. The problem addressed is accurately identifying the orientation of an object during motion, especially when vertical acceleration varies due to factors like terrain or movement patterns. The method involves acquiring vertical acceleration data over time, identifying successive vertical acceleration minima below a specified threshold, and using orientation data captured at the exact moments these minima occur. This approach ensures that orientation measurements are taken during consistent, repeatable conditions, improving accuracy. The method may also include filtering the acceleration data to remove noise and selecting a subset of minima based on predefined criteria, such as time intervals or amplitude thresholds. By correlating orientation data with these specific acceleration events, the system can reliably track changes in orientation over time, even in dynamic environments. The technique is useful in applications like navigation, robotics, and motion analysis where precise orientation data is critical.
12. The method as recited in claim 1 , further comprising initial determination of the step length by: (g) carrying the imaging device while walking along a substantially straight path having a known length; (h) repeatedly acquiring vertical acceleration data during walking along the straight path using the inertial measurement unit; (i) counting the number of walking steps taken during walking along the straight path by detecting a repeating characteristic of the vertical acceleration data; (j) calculating the step length during the walk along the straight path by dividing the known length by the number of walking steps counted in step (i); and (k) storing the step length in a non-transitory tangible computer-readable storage medium, wherein steps (g) through (k) are performed prior to step (a).
The invention relates to a method for determining step length using an imaging device equipped with an inertial measurement unit (IMU). The method addresses the challenge of accurately measuring step length for applications such as indoor navigation or motion tracking, where precise step length data is essential for accurate positioning or activity monitoring. The method involves an initial calibration process to determine the user's step length. This is done by having the user walk along a straight path of known length while carrying the imaging device. During this walk, the IMU repeatedly captures vertical acceleration data. The system then analyzes this data to detect and count the number of walking steps by identifying a repeating characteristic in the vertical acceleration pattern. The step length is calculated by dividing the known path length by the counted number of steps. This calculated step length is stored in a non-volatile memory for later use in subsequent steps of the method. This calibration step ensures that the step length used in later applications is accurate and tailored to the individual user, improving the reliability of any subsequent motion tracking or navigation functions performed by the device. The method leverages inertial sensing to automate step counting and eliminate manual measurements, enhancing convenience and precision.
13. The method as recited in claim 1 , wherein steps (b)-(f) and (h)-(j) are performed by the imaging device.
This invention relates to an imaging system that automates the capture, processing, and transmission of images. The system addresses the problem of inefficient image handling in devices that require manual intervention for tasks such as image capture, enhancement, and transmission. The imaging device itself performs key steps in the process, reducing reliance on external systems or user input. The method involves capturing an image using an imaging device, such as a camera or scanner. The device then processes the image to enhance its quality, which may include adjusting brightness, contrast, or sharpness. The processed image is stored locally on the device. The device also generates metadata associated with the image, such as timestamps, device settings, or location data. This metadata is stored alongside the image. The device then compresses the image to optimize storage or transmission efficiency. The compressed image and metadata are transmitted to a remote server or another device for further use. The imaging device also performs error checking to ensure data integrity during transmission. If an error is detected, the device retransmits the data until successful. The system ensures that the imaging device handles these steps autonomously, minimizing the need for external processing or user interaction.
14. The method as recited in claim 1 , wherein steps (d), (e), (i), and (j) are performed by a computer system which receives the orientation data and the vertical acceleration data from the imaging device.
A system and method for analyzing motion and orientation data from an imaging device, such as a camera or sensor, to detect and correct movement artifacts in captured images or video. The technology addresses the problem of image distortion caused by unintended device motion, such as shaking or tilting, which degrades visual quality and accuracy in applications like surveillance, medical imaging, or augmented reality. The method involves capturing orientation data (e.g., angular position) and vertical acceleration data from the imaging device, then processing this data to compensate for motion-induced distortions. A computer system performs key steps, including receiving the orientation and acceleration data, analyzing it to determine motion patterns, and applying corrective adjustments to stabilize the captured images. The system may also compare the motion data against predefined thresholds to trigger stabilization algorithms or alert operators to excessive movement. By integrating motion sensing and real-time correction, the invention improves image stability and usability in dynamic environments. The method is particularly useful in scenarios where precise alignment or motion tracking is critical, such as in robotics, autonomous navigation, or high-precision imaging systems.
15. A method for identifying a part of a target object, comprising: (a) carrying an imaging device from an initial location that is known in a frame of reference of the target object to a subsequent location that is not known in the frame of reference of the target object while walking along a path, wherein an area on the target object including a part of interest is in a field-of-view of the imaging device when the imaging device is at the subsequent location; (b) executing a dead-reckoning algorithm that combines orientation data from an inertial measurement unit incorporated in the imaging device with walking step information to produce a piecewise linear approximation of the path for measurement of motion of the imaging device relative to the initial location; (c) calculating a current position and a current orientation for the subsequent location of the imaging device relative to the frame of reference of the target object using the measurement of the motion of the imaging device relative to the initial location; (d) displaying a three-dimensional model image of at least a portion of the target object with a viewpoint that is a function of the calculated current position and current orientation of the imaging device; (e) selecting a point in a portion of the displayed three-dimensional model image that depicts the part of interest; (f) retrieving part identification data from a three-dimensional model database corresponding to the part of interest; and (g) displaying alphanumeric symbology that depicts the identity of the part of interest contained in the retrieved part identification data.
This invention relates to a method for identifying parts of a target object using an imaging device and dead-reckoning navigation. The problem addressed is accurately determining the position and orientation of a handheld imaging device relative to a known reference frame of a target object, even when the device's subsequent location is unknown, to enable precise part identification. The method involves carrying an imaging device along a path from an initial known location to a subsequent unknown location, where the part of interest is within the device's field of view. An inertial measurement unit (IMU) in the device provides orientation data, while walking step information is combined with the IMU data using a dead-reckoning algorithm to estimate the device's motion relative to the initial location. This produces a piecewise linear approximation of the path, allowing calculation of the device's current position and orientation in the target object's reference frame. A three-dimensional model of the target object is then displayed with a viewpoint adjusted based on the calculated position and orientation. A user selects a point in the displayed model corresponding to the part of interest, and part identification data is retrieved from a database. The part's identity is then displayed as alphanumeric symbology, enabling accurate part recognition. This approach is useful in applications requiring precise part identification in dynamic or unstructured environments.
16. The method as recited in claim 15 , further comprising: acquiring feature image data while the area on the target object including the part of interest is in the field-of-view of the imaging device; displaying a feature image corresponding to the acquired feature image data as an overlay or background to the three-dimensional model image; and adjusting the viewpoint until the three-dimensional model image being displayed matches the feature image.
This invention relates to a method for aligning a three-dimensional (3D) model of a target object with a feature image of the object, improving accuracy in visualizing and analyzing specific parts of interest. The method addresses the challenge of precisely matching a 3D model with real-world imagery, which is critical in applications like medical imaging, industrial inspection, and augmented reality. The method involves acquiring feature image data of a target object while a part of interest is within the field-of-view of an imaging device. The feature image, derived from this data, is displayed as an overlay or background to a pre-existing 3D model image of the object. The user then adjusts the viewpoint of the 3D model until it aligns with the feature image, ensuring accurate spatial correspondence between the model and the real-world features. The method may also include generating the 3D model by capturing multiple images of the target object from different viewpoints and processing these images to construct the model. Additionally, the method may involve tracking the position and orientation of the imaging device to facilitate alignment. The feature image can be adjusted in real-time to enhance the matching process, ensuring precise visualization of the part of interest. This approach improves the accuracy of 3D model alignment with real-world imagery, enabling more reliable analysis and decision-making in various technical and medical applications.
17. The method as recited in claim 15 , further comprising acquiring vertical acceleration data during walking along the path using the inertial measurement unit, wherein executing the dead-reckoning algorithm comprises counting a number of walking steps taken during walking along the path by detecting a repeating characteristic of the vertical acceleration data.
This invention relates to a navigation system for determining a user's position along a path, particularly in environments where global navigation satellite systems (GNSS) are unreliable or unavailable. The system addresses the challenge of maintaining accurate positioning in such conditions by combining inertial measurement unit (IMU) data with dead-reckoning techniques to estimate the user's location. The method involves acquiring IMU data, including vertical acceleration, while the user walks along a path. The dead-reckoning algorithm processes this data to count the number of walking steps by detecting a repeating pattern in the vertical acceleration, which corresponds to each step. This step-counting feature enhances the accuracy of the dead-reckoning algorithm by providing a more precise measure of distance traveled. The system may also use additional IMU data, such as angular velocity and linear acceleration, to refine the position estimate further. The method is particularly useful for indoor navigation, urban canyons, or other GNSS-denied environments where traditional positioning methods fail. By leveraging IMU data and step detection, the system provides a reliable way to track movement without external references.
18. The method as recited in claim 15 , wherein steps (b) and (c) are performed by a processor incorporated in the imaging device.
The invention relates to image processing systems, specifically methods for enhancing image quality in imaging devices. The problem addressed is the computational inefficiency and latency in processing images, particularly in real-time applications where image quality must be improved without significant delay. Traditional systems often rely on external processors, which introduce latency and increase system complexity. The method involves capturing an image using an imaging device, such as a camera or scanner, and then processing the image to enhance its quality. The processing steps include analyzing the image data to identify areas requiring correction, such as noise reduction, contrast adjustment, or sharpness enhancement. The corrected image data is then output for display or storage. A key innovation is that the processing steps are performed by a processor integrated directly into the imaging device, eliminating the need for external processing hardware. This integration reduces latency, improves efficiency, and simplifies system design. The method may also include additional steps, such as pre-processing the image before analysis or post-processing the corrected image for further refinement. The processor in the imaging device executes these steps, ensuring real-time performance and minimizing data transfer delays. This approach is particularly useful in applications requiring rapid image processing, such as medical imaging, surveillance, or industrial inspection.
19. The method as recited in claim 15 , wherein steps (b) and (c) are performed by a computer system external to the imaging device.
A method for processing imaging data involves capturing an image using an imaging device and analyzing the image to detect a target object. The method includes transmitting the captured image to a computer system external to the imaging device, where the computer system performs the analysis to identify the target object. The external computer system may use machine learning algorithms or other image processing techniques to detect the target object within the image. The results of the analysis, such as the location or characteristics of the detected object, are then transmitted back to the imaging device or another system for further processing or display. This approach offloads computationally intensive tasks from the imaging device to a more powerful external system, improving efficiency and accuracy. The method may be applied in various fields, including surveillance, medical imaging, or industrial inspection, where real-time or high-precision object detection is required. The external computer system may also store the analyzed data for future reference or additional processing.
20. An apparatus comprising a digital camera, an inertial measurement unit, a display screen, a non-transitory tangible computer-readable storage medium storing three-dimensional model data of a target object, a three-dimensional visualization software application, and a dead-reckoning algorithm, and a processor configured to perform the following operations: (a) counting a number of walking steps taken by a user carrying the apparatus during walking along a path from an initial location to a subsequent location by detecting a repeating characteristic of vertical acceleration data acquired by the inertial measurement unit; (b) calculating a current position and a current orientation for the subsequent location of the digital camera relative to a frame of reference of the target object using the dead-reckoning algorithm, orientation data acquired by the inertial measurement unit, the number of walking steps, and a step length; (c) retrieving the three-dimensional model data of the target object from the non-transitory tangible computer-readable storage medium; and (d) using the three-dimensional visualization software application to control the display screen to display a three-dimensional model image of at least a portion of the target object with a viewpoint that is a function of the calculated current position and current orientation of the imaging device, the three-dimensional model image depicting three-dimensional model data retrieved from a three-dimensional model database.
This invention relates to a portable apparatus for augmented reality navigation and visualization. The apparatus includes a digital camera, an inertial measurement unit (IMU), a display screen, and a processor. The processor executes a dead-reckoning algorithm to track a user's movement by counting walking steps via vertical acceleration data from the IMU. The system calculates the user's current position and orientation relative to a target object using step count, step length, and IMU orientation data. The apparatus also stores three-dimensional (3D) model data of the target object and a 3D visualization software application. The processor retrieves the 3D model data and renders a 3D model image on the display screen, adjusting the viewpoint based on the user's calculated position and orientation. This allows the user to visualize the target object in 3D as they move, providing spatial context and navigation assistance. The system is designed for applications where real-time 3D visualization of a known object is needed, such as indoor navigation, maintenance, or inspection tasks. The apparatus combines dead-reckoning with 3D modeling to enhance situational awareness without requiring external positioning systems.
21. The apparatus as recited in claim 20 , wherein the processor is further configured to use the three-dimensional visualization software application to control the display screen to display a feature image of the same portion of the target object as an overlay or background to the three-dimensional model image, the feature image representing at least portions of feature image data captured by the digital camera while the apparatus is at the subsequent location.
This invention relates to a system for visualizing and analyzing three-dimensional (3D) models of objects, particularly for applications in inspection, measurement, or quality control. The system addresses the challenge of accurately aligning and comparing 3D model data with real-world images to verify features, detect defects, or guide inspections. The apparatus includes a processor, a display screen, and a digital camera. The processor executes 3D visualization software to generate and display a 3D model image of a target object. The system is capable of capturing feature image data of the target object using the digital camera at a subsequent location, where the camera's position may differ from the original 3D scanning position. The processor then overlays or underlays a feature image derived from this captured data onto the 3D model image, ensuring alignment with the same portion of the object. This allows users to compare the 3D model with real-world visual data, facilitating tasks such as defect detection, dimensional verification, or alignment validation. The feature image may represent surface details, markings, or other visual characteristics captured by the camera, providing a real-time or post-processing reference for analysis. The system enhances accuracy in inspections by integrating dynamic visual data with pre-existing 3D models, improving decision-making in quality control and manufacturing processes.
22. The apparatus as recited in claim 20 , wherein operation (b) comprises computing an offset transformation matrix that represents position and orientation offsets of the subsequent location relative to the initial location.
This invention relates to a system for determining positional and orientational offsets between locations in a spatial environment. The problem addressed is accurately tracking changes in position and orientation between an initial location and a subsequent location, which is critical for applications such as robotics, augmented reality, and autonomous navigation. The apparatus includes a sensor system that captures spatial data at an initial location and a subsequent location. The system processes this data to compute a transformation matrix that quantifies the positional and orientational differences between the two locations. The transformation matrix is derived from the spatial data, allowing precise calculation of translation and rotation offsets. This enables the apparatus to determine how the subsequent location differs from the initial location in terms of both position and orientation, which is essential for tasks requiring spatial awareness and movement tracking. The computed transformation matrix can be used to adjust navigation paths, update positional references, or align spatial data in real-time applications. The invention improves accuracy in spatial tracking by leveraging precise transformation calculations, reducing errors in position and orientation determination.
23. The method as recited in claim 6 , further comprising: using Fast Fourier Transform analysis of the acceleration data to compute an updated estimate of gait cycle step frequency; and refining a mathematical model of the dead-reckoning process using a walk ratio and the updated estimate of gait cycle step frequency.
This invention relates to improving dead-reckoning navigation systems, particularly for pedestrian or gait-based tracking. Dead-reckoning systems estimate position by tracking movement from a known starting point, but errors accumulate over time due to inaccuracies in step detection and gait modeling. The invention addresses this by refining the dead-reckoning process using gait cycle analysis. The method involves analyzing acceleration data from a wearable device to compute an updated estimate of gait cycle step frequency using Fast Fourier Transform (FFT) analysis. This frequency estimate is then used to refine a mathematical model of the dead-reckoning process. The refinement incorporates a walk ratio, which relates step length to step frequency, to improve position accuracy. By dynamically adjusting the model based on real-time gait data, the system reduces cumulative errors and enhances tracking precision. The FFT analysis extracts frequency components from acceleration signals, identifying the dominant frequency corresponding to the gait cycle. The walk ratio, derived from empirical or user-specific data, scales step length estimates to match the observed step frequency. This adaptive approach compensates for variations in walking speed, stride length, and terrain, making the system more robust for real-world applications. The refined model updates the dead-reckoning calculations iteratively, ensuring continuous improvement in position estimation. This technique is particularly useful in environments where GPS signals are unreliable, such as indoors or in urban canyons.
24. The method as recited in claim 23 , further comprising: detecting a deviation of the user's actual step length from the step length used in step (a); and displaying symbology on a display screen of the imaging device to indicate an amount of the deviation.
This invention relates to a method for improving user posture and movement accuracy during imaging procedures, particularly in medical or industrial applications where precise positioning is critical. The method involves tracking a user's step length during movement and comparing it to a predefined or ideal step length to ensure consistent and accurate positioning. The system detects deviations between the user's actual step length and the intended step length, then provides visual feedback via symbology on a display screen to alert the user of any discrepancies. This feedback helps the user correct their movement in real-time, reducing errors in positioning and improving the accuracy of the imaging or procedural outcome. The method may also include additional steps such as adjusting the imaging device's field of view or calibration parameters based on the detected deviations to further enhance precision. The invention is particularly useful in applications where small positional errors can significantly impact results, such as in medical imaging, robotic surgery, or industrial inspection. By providing immediate feedback, the system ensures that the user maintains optimal movement patterns, leading to more reliable and consistent outcomes.
Unknown
October 13, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.